fix(qoder): report usage to all clients and stop inlining large attachments

- Coalesce Qoder's empty finish-in-delta frame with the later choices:[] usage
  frame so OpenAI and Claude clients receive prompt_tokens, completion_tokens
  and cache-hit tokens (the dashboard already saw them)
- Upload inlined images through /api/v2/image/upload like qodercli, and stub
  oversized non-image files instead of stuffing 30MB+ data URIs into
  agent_chat_generation
- Emit response.completed -> response.usage for chat-native upstreams so
  /v1/responses clients (Codex CLI, sub2api) no longer log 0/0/0
- Keep Claude message_delta.usage working when usage arrives without choices[0]
- Escalate to the smallest advertised Qoder context tier (200K/400K/1M) when
  the estimated prompt no longer fits max_input_tokens
- Pass apiKey for PAT connections and list hidden enable:false catalog keys
  from /v1/models

tests/unit/openai-to-claude.test.js

@@ -203,4 +203,31 @@ describe("openaiToClaudeResponse", () => {
       limit: 120
     });
   });
+
+  it("records usage from a choices:[] frame so the finish chunk can emit it", () => {
+    const state = { toolCalls: new Map() };
+    expect(openaiToClaudeResponse({
+      usage: {
+        prompt_tokens: 90,
+        completion_tokens: 7,
+        prompt_tokens_details: { cached_tokens: 30 },
+      },
+      choices: [],
+    }, state)).toBeNull();
+    expect(state.usage).toEqual({
+      input_tokens: 60,
+      output_tokens: 7,
+      cache_read_input_tokens: 30,
+    });
+
+    const events = openaiToClaudeResponse({
+      id: "chatcmpl-qoder-finish",
+      model: "qoder/auto",
+      choices: [{ index: 0, delta: {}, finish_reason: "stop" }],
+    }, state);
+    const delta = events.find((e) => e.type === "message_delta");
+    expect(delta.usage.input_tokens).toBe(60);
+    expect(delta.usage.output_tokens).toBe(7);
+    expect(delta.usage.cache_read_input_tokens).toBe(30);
+  });
 });